Slice-selective Inversion Research Articles

Accelerated 2D radial Look-Locker T1 mapping using a deep learning-based rapid inversion recovery sampling technique.

Efficient abdominal coverage with T1-mapping methods currently available in the clinic is limited by the breath hold period (BHP) and the time needed for T1 recovery. This work develops a T1-mapping framework for efficient abdominal coverage based on rapid T1 recovery curve (T1RC) sampling, slice-selective inversion, optimized slice interleaving, and a convolutional neural network (CNN)-based T1 estimation. The effect of reducing the T1RC sampling was evaluated by comparing T1 estimates from T1RC ranging from 0.63 to 2.0s with reference T1 values obtained from T1RC =2.5-5s. Slice interleaving methodologies were evaluated by comparing the T1 variation in abdominal organs across slices. The repeatability of the proposed framework was demonstrated by performing acquisition on test subjects across imaging sessions. Analysis of in vivo data based on retrospectively shortening the T1RC showed that with the CNN framework, a T1RC =0.84 s yielded T1 estimates without significant changes in mean T1 (p> 0.05) or significant increase in T1 variability (p> 0.48) compared to the reference. Prospectively acquired data using T1RC =0.84 s, an optimized slice interleaving scheme, and the CNN framework enabled 21 slices in a 20 s BHP. Analyses across abdominal organs produced T1 values within 2% of the reference. Repeatability experiments yielded Pearson's correlation, repeatability coefficient, and coefficient of variation of 0.99, 2.5%, and 0.12%, respectively. The proposed T1 mapping framework provides full abdominal coverage within a single BHP.

Hadamard-encoded dual-voxel SPECIAL: Short-TE MRS acquired in two brain regions simultaneously using Hadamard encoding.

The spin-echo, full-intensity acquired localized (SPECIAL) sequence is a method for single-voxel, localized MRS in vivo with short TEs. In this study we modified the SPECIAL sequence to simultaneously record spectra from two volumes of interest. This new technique is called Hadamard-encoded dual-voxel SPECIAL (HD-SPECIAL). The SPECIAL sequence consists of a spin echo localized to a column of tissue, preceded by a slice-selective inversion pulse in alternating scans to invert a section of the column. Full localization is achieved by subtraction of the inversion-on scans from the inversion-off scans. In HD-SPECIAL, the two-step inversion scheme is replaced by a four-step Hadamard-encoded scheme involving single-band and dual-band inversion pulses to select two regions of the spin-echo column. By appropriate Hadamard combination of the four acquired shots, spectra can be reconstructed from both desired regions. This approach does not rely on parallel imaging reconstruction. Using a 3T scanner, HD-SPECIAL localization is demonstrated both in phantoms and in the human brain in vivo, and the performance of HD-SPECIAL is assessed by comparing with the conventional SPECIAL sequence. Phantom and in vivo measurements show excellent agreement between measures from HD-SPECIAL and SPECIAL sequences. Relative to consecutive SPECIAL measurements from two regions, HD-SPECIAL reduces the total scan time 2-fold with minimal penalty in terms of spectral quality or SNR. The HD-SPECIAL sequence enables reliable acquisition of MR spectra simultaneously from two regions at 3 T, offering the potential to study interregional variations in metabolite concentrations.

Multiphoton magnetic resonance in imaging: A classical description and implementation.

To develop a classical geometric interpretation of multiphoton excitation and apply it to MRI. To investigate ways in which multiphoton excitation can enable novel imaging techniques. We present a fully geometric view of multiphoton excitation by taking a particular rotating frame transformation. In this rotating frame, we find that multiphoton excitations appear just like single-photon excitations again, and therefore, we can readily generalize concepts already explored in standard single-photon excitation. With a homebuilt low frequency coil, we execute a standard slice selective pulse sequence with all of its excitations replaced by their equivalent two-photon versions. In the case of no extra hardware, we use oscillating gradients as a source of extra photons for excitation. Finally, with the multiphoton interpretation of oscillating gradients, we present a novel way to transform a standard slice selective adiabatic inversion pulse into a multiband version without modifying the RF pulse itself. The addition of oscillating gradients creates multiphoton resonances at multiple spatial locations and allows for adiabatic inversions at each location. With Bloch-Siegert shift corrections, analytical multiphoton excitation expressions match with Bloch equation simulations. Two-photon gradient-echo images of a lemon and a pork rib match with their single-photon counterparts. Frequency-offset RF combined with oscillating gradients generate excitation where the RF alone does not. The multiphoton interpretation presents new flexibilities for imaging. Excitation needs not be bound to the Larmor frequency, which opens doors to RF pulse design beyond the usual filter design and the potential for further imaging innovations.

A new sequence for shaped voxel spectroscopy in the human brain using 2D spatially selective excitation and parallel transmission.

Spatially selective excitation in two dimensions (2D-SSE) utilizing parallel transmission was applied as a means to acquire signal from voxels adapted to the anatomy of interest for in vivo (1) H MR spectroscopy. A novel method to select spectroscopy voxels with arbitrary shapes in two dimensions was investigated. An on-off scheme with an adiabatic slice selective inversion pulse preceding a 2D-SSE pulse together with a segmented inward spiral excitation k-space trajectory enabled rapid free induction decay acquisitions. Performance of the sequence was evaluated in simulations, phantom experiments, and in vivo measurements at 3 T. High spatial fidelity of the excitation profile was achieved for different target shapes and with little off-resonance deterioration. Metabolite concentrations in human brain determined with the new sequence were quantified with Cramér-Rao lower bounds less than 20%. They were in the physiological range and did not deviate systematically from results acquired with a conventional SPECIAL sequence. In conclusion, a new approach for shaped voxel MRS in the human brain is presented, which complements existing sequences. Simulations show that 2D-SSE pulses yield reduced chemical shift artifact when compared with conventional localization methods. Copyright © 2016 John Wiley & Sons, Ltd.

Myocardial perfusion quantification using theT1-based FAIR-ASL method: The influence of heart anatomy, cardiopulmonary blood flow and look-locker readout

The quantification of myocardial perfusion using a Look-Locker flow-sensitive alternating inversion recovery- arterial spin labeling experiment is considered. Due to the anatomy of the heart, a substantial but unintended partial inversion of the inflowing blood occurs during the slice-selective inversion. Both, the partial inversion as well as the Look-Locker pulse train, influence the myocardial perfusion quantification and are addressed in this work. The mean relaxation time approximation is used to calculate the monoexponential relaxation time of the signal in perfused tissue under Look-Locker readout. The left ventricular blood serves as an approximation of the inflowing blood in the description of FAIR-ASL measurements with global and slice-selective inversion to correctly quantify the myocardial perfusion. The analysis shows that the myocardial perfusion can be overestimated if the T1 -based quantification method is not adapted respecting the Look-Locker pulse train explicitly. Additionally, it turns out that without correction for the partial inversion of the blood pool during the slice-selective inversion the myocardial perfusion is underestimated. It is shown that the Look-Locker readout as well as the nonideal slice-selective inversion experiment have a considerable influence and have to be included properly to correctly quantify myocardial perfusion.

Multislice cardiac arterial spin labeling using improved myocardial perfusion quantification with simultaneously measured blood pool input function

Myocardial blood flow (MBF) is an important indicator of cardiac tissue health, which can be measured using arterial spin labeling. This study aimed to develop a new method of MBF quantification with blood pool magnetization measurement ("bpMBF quantification") that allows multislice cardiac arterial spin labeling. A multislice segmented ECG-gated Look-Locker T1 mapping sequence was validated. Quantification of multislice arterial spin labeling is not straightforward due to the large volume of blood inverted following slice-selective inversion. For bpMBF quantification, a direct measurement of the left-ventricle blood pool magnetization was used to approximate the blood input function into the Bloch equations. Simulations and in vivo measurements in the mouse heart were performed to evaluate the bpMBF method. Measurements indicated that blood pool magnetization requires ∼3 s to return to equilibrium following slice-selective inversion. Simulation and in vivo results show that bpMBF quantification is robust to variations in slice-selective thickness and therefore applicable to multislice acquisition, whereas traditional methods are likely to underestimate multislice perfusion. In vivo, single and multislice perfusion values matched well when quantified using bpMBF. The first multislice cardiac arterial spin labeling technique has been presented, which can be used for accurate perfusion measurements in studies of cardiac disease.

Oxygen-enhanced lung magnetic resonance imaging: influence of inversion pulse slice selectivity on inversion recovery half-Fourier single-shot turbo spin-echo signal

The purpose of this study was to evaluate in vivo the influence of inversion pulse slice selectivity on oxygen-enhanced magnetic resonance imaging (MRI). Thirteen healthy volunteers were studied with a two-dimensional cardiac- and respiratory-gated adiabatic inversion-recovery half-Fourier single-shot turbo spin-echo (HASTE) sequence with either slice-selective or non-slice-selective inversion recovery (IR) pulse at inversion times increasing from 300 to 1400 ms. The signal-to-noise ratio (SNR) at every inversion time (TI), real signal difference (ΔSI), and relative enhancement ratio of lung parenchyma at TI ≥ 800 ms were statistically compared for oxygen-enhanced and non-oxygen-enhanced MR images with slice-selective or non-slice-selective IR pulses. The SNRs of acquisitions with slice-selective IR pulses were significantly higher than those of non-slice-selective IR pulses (P < 0.05). At TI 800 ms, the ΔSI of lung parenchyma on IR-HASTE images with slice-selective inversion pulse type was significantly higher than on that with the non-slice-selective type (P < 0.05). Relative enhancement ratios of the slice-selective IR pulses were significantly lower than those of non-slice-selective IR pulses at TIs between 800 and 1400 ms (P < 0.05). Slice selectivity of inversion pulse type affects oxygen-enhanced MRI in vivo.

Improved cardiac arterial spin labelling in the mouse heart by optimisation of acquisition and analysis

Methods Perfusion maps are generated by pixel-wise comparison of T1 maps from slice-selective and global inversion recoveries. T1 was quantified using ECG-gated LookLocker approach [3], with 4 lines of k-space acquired every heartbeat to speed up acquisition by approximately four times compared with previous approaches (eg. [2],[3]). A data logger (CED) was used to record ECG, RF and respiration events and was programmed to gate on each ECG peak detected as well as to interpolate peaks missed by physiological monitoring software. Using the information recorded, lines of k-space were binned as ‘acquired during respiration’ or not. Images with many lines or centre lines of k-space acquired during respiration were rejected. ASL data were acquired on a 9.4T Varian system with 35mm-long volume RF coil (Rapid) (TE=1.18ms, flip=5°, resolution=200μm, slice thickness=1.5mm) and image analysis was performed in Matlab.

Comparison of selective arterial spin labeling using 1D and 2D tagging RF pulses

Generic arterial spin labeling (ASL) techniques label all brain feeding arteries. In this work, we used two different selective ASL (SASL) methods to show the perfusion of one single artery. A slice selective inversion of an area including the desired vessel was compared to a multidimensional RF pulse with Gaussian profile to label only the artery of interest. Perfusion images with a resolution of 2 x 2 x 5 mm 3 are shown that were acquired after tagging only the internal carotid artery of healthy volunteers. In addition, both techniques were applied to a patient with an extra-intracranial bypass to illustrate its perfusion territory. These perfusion images are consistent with a standard angiography. SASL imaging with a resolution of 2 x 2 x 5 mm 3 is possible in a total scan time of 5 min. The presented MR techniques may in part replace the assessment of revascularization success by conventional angiography.

Tailored RF pulse for magnetization inversion at ultrahigh field

The radiofrequency (RF) transmit field is severely inhomogeneous at ultrahigh field due to both RF penetration and RF coil design issues. This particularly impairs image quality for sequences that use inversion pulses such as magnetization prepared rapid acquisition gradient echo and limits the use of quantitative arterial spin labeling sequences such as flow-attenuated inversion recovery. Here we have used a search algorithm to produce inversion pulses tailored to take into account the heterogeneity of the RF transmit field at 7 T. This created a slice selective inversion pulse that worked well (good slice profile and uniform inversion) over the range of RF amplitudes typically obtained in the head at 7 T while still maintaining an experimentally achievable pulse length and pulse amplitude in the brain at 7 T. The pulses used were based on the frequency offset correction inversion technique, as well as time dilation of functions, but the RF amplitude, frequency sweep, and gradient functions were all generated using a genetic algorithm with an evaluation function that took into account both the desired inversion profile and the transmit field inhomogeneity.

A free-breathing non-contrast-enhanced pulmonary magnetic resonance angiography at 3 Tesla.

The breathhold contrast-enhanced three-dimensional magnetic resonance angiography (MRA) using T1-weighted gradient-echo imaging sequence is the standard technique for MRA of the thorax. However, this technique is not desirable for certain patients with respiratory insufficiency, serious renal impairment, or allergy to contrast agents. The objective of this study was to optimize and evaluate a non-contrast-enhanced free-breathing pulmonary MRA protocol at 3 Tesla. The time-of-flight protocol was based on a two-dimensional T1-weighted turbo field echo sequence with slice-selective inversion recovery and magnetization transfer preparation together with respiratory navigator gating, cardiac gating, and parallel imaging. Optimal values for time of inversion delay, flip angle and slice thickness were experimentally determined and used for all subjects. Excellent pulmonary MRA images, in which the 7th order branches of pulmonary arteries could be reliably identified, were obtained in the 12 free-breathing healthy volunteers. TI of approximately 300 ms provides the best suppression of background thoracic and cardiac muscles and effective inflow enhancement. With increasing flip angle, the pulmonary vessels gradually brightened and exhibited optimal contrast at 20 degrees-30 degrees. The 2 mm slice thickness and 0.5 mm slice overlap is suitable for visualization of the peripheral pulmonary vessel. The MRA protocol at 3 Tesla may have clinical significance for pulmonary vascular imaging in patients who are not available for contrast-enhanced 3D MRA and CT angiography examination or are unable to sustain a long breath-hold.

Quantification of slow flow using FAIR

Phase contrast (PC)-based MRI methods are considered to be the most accurate approach for spatially resolved flow quantification, but the measurement of very slow velocities requires signal detection at long echo times and the application of strong field gradients. On the other hand, measurements based on time-of-flight or inflow effects can be conducted at short echo times and without flow-encoding gradients. A method for imaging flow at velocities of the order of 0.1 mm/s is presented and validated here. It consists of measuring the apparent spin-lattice relation rate (R(1)*) of the flowing fluid using magnetization preparation by alternating slice-selective and nonselective inversion pulses (FAIR or flow-sensitive alternating inversion recovery) and a fast gradient-echo detection sequence. This method is appropriate for the quantitative imaging of slow flow at low Reynolds numbers in fluids where the T(2) values are too short to allow sensitive flow measurements by phase contrast-based methods.

Fast oxygen‐enhanced multislice imaging of the lung using parallel acquisition techniques

The purpose of this study was to evaluate an optimized multislice acquisition technique for oxygen-enhanced MRI of the lung using slice-selective inversion and refocusing pulses in combination with parallel imaging. An inversion recovery HASTE sequence was implemented with respiratory triggering to perform imaging in end-expiration and with ECG triggering to avoid image acquisition during the systolic phase. Inversion pulses and the readout of echo trains could be interleaved to decrease acquisition time. The sequence was evaluated in 15 healthy volunteers, comparing three acquisition schemes: (1) acquisition of four slices without parallel imaging; (2) acquisition of four slices with parallel imaging; (3) acquisition of six slices with parallel imaging. These multislice acquisitions were repeated 80 times with alternating inhalation of room air and oxygen. The oxygen-induced signal increase showed no significant difference with and without parallel imaging. However, only with parallel imaging did the interleaved acquisition of six or more slices become possible, thus enabling a more complete anatomic coverage of the lung. The average required end-expiration time per repetition to acquire six slices could be significantly reduced from 4112 ms without to 2727 ms with parallel imaging. Total acquisition time varied between 8 and 13 min depending on the respiratory frequency.

Accurate myocardial T1 measurements: Toward quantification of myocardial blood flow with arterial spin labeling

In this study, we investigated a method for accurately measuring myocardial T(1) for the quantification of myocardial blood flow (MBF) with arterial spin labeling (ASL). A single-shot gradient-echo (GE)-based ASL sequence with an adiabatic hyperbolic secant inversion recovery pulse was modified to acquire a pair of myocardial T(1)'s within a breath-hold. A multivariable regression algorithm that accounted for the magnetization saturation effects was developed to calculate T(1). The MBF was then determined with a well-developed model. The accuracy of our T(1) calculation was first evaluated in a phantom, and then in six dogs for the MBF calculation, with (N = 4) and without (N = 2) coronary artery stenosis. In the phantom study, the accuracy of T(1) measured with a slice-selective inversion prepared pulse was within 2.5% of error. In healthy dogs, the MBF increased 2-5 times during vasodilation. In contrast, regional differences of MBF were well visualized in the stenotic dogs during vasodilation (perfusion reserve of 2.75 +/- 0.83 in normal myocardium, and 1.46 +/- 0.75 in the stenotic area). A correlation analysis revealed a close agreement in MBF between the ASL and microsphere (MS) in both healthy and stenotic dogs. In summary, the modified ASL technique and T(1) regression algorithm proposed here provide an accurate measurement of myocardial T(1) and demonstrate potential for reliably assessing MBF at steady state.

Inversion Prepared Coronary MR Angiography: Direct Visualization of Coronary Blood Flow

Visualization of coronary blood flow by means of a slice-selective inversion pre-pulse in concert with bright-blood coronary MRA. Coronary magnetic resonance angiography (MRA) of the right coronary artery (RCA) was performed in eight healthy adult subjects on a 1.5 Tesla MR system (Gyroscan ACS-NT, Philips Medical Systems, Best, NL) using a free-breathing navigator-gated and cardiac-triggered 3D steady-state free-precession (SSFP) sequence with radial k-space sampling. Imaging was performed with and without a slice-selective inversion pre-pulse, which was positioned along the main axis of the coronary artery but perpendicular to the imaging volume. Objective image quality parameters such as SNR, CNR, maximal visible vessel length, and vessel border definition were analyzed. In contrast to conventional bright-blood 3D coronary MRA, the selective inversion pre-pulse provided a direct measure of coronary blood flow. In addition, CNR between the RCA and right ventricular blood pool was increased and the vessels had a tendency towards better delineation. Blood SNR and CNR between right coronary blood and epicardial fat were comparable in both sequences. The combination of a free-breathing navigator-gated and cardiac-triggered 3D SSFP sequence with a slice-selective inversion pre-pulse allows for direct and directional visualization of coronary blood flow with the additional benefit of improved contrast between coronary and right ventricular blood pool.

Multislice double inversion-recovery black-blood imaging with simultaneous slice reinversion.

To develop a technique for time-efficient multislice double inversion-recovery (DIR) black-blood imaging and to test its applicability and limitations for high-resolution imaging of carotid arteries. A multislice DIR pulse sequence with fast spin-echo (FSE) readout was implemented on a 1.5 T magnetic resonance (MR) scanner. The principle of the method is that a slice-selective inversion in a DIR preparation reinverts an entire slice pack, and all slices are imaged within repetition time (TR). The number of slices acquired per TR (N) controls the inversion time (TI) to execute the readout for each slice at the zero-crossing point of blood. Multislice DIR images (TR/TE = 2500/9 msec) of carotid arteries were obtained with variable N = 2-8 from four subjects. The method was compared with the standard single-slice DIR and inflow saturation techniques. Multislice DIR with N = 2-6 provided similar flow suppression in carotid arteries as single-slice DIR. At all N = 1-8, blood suppression by DIR was significantly better than by inflow saturation. An additional limitation of multislice DIR was saturation of the signal from stationary tissues that worsened visualization of the vessel wall at N >or= 6. Multislice DIR provides up to eight-fold improvement of time-efficiency relative to single-slice DIR and high-quality blood suppression.

Adiabatic radiofrequency pulse forms in biomedical nuclear magnetic resonance

Adiabatic radio frequency (RF) pulses are in widespread use in biomedical magnetic resonance imaging and spectroscopy. The primary advantage of adiabatic pulses is that provided the condition for adiabatcity is satisfied they can be made insensitive to inhomogeneities in the RF field. In this pedagogical article the principles of adiabatic fast passage (AFP) are explained, and the use of AFP to invert both stationary and flowing spin systems is examined. The hyperbolic secant pulse is presented as a pulse capable of performing slice selective adiabatic inversion. Lower power alternatives to this pulse are described, and the principle of offset independent constant adiabaticity is elucidated. Instantaneous reversal of the orientation of the effective RF field as a means of producing excitation and refocusing pulses is presented, as are methods of obtaining slice selective excitation with these pulses.

Improved perfusion quantification in FAIR imaging by offset correction.

Perfusion quantification using pulsed arterial spin labeling has been shown to be sensitive to the RF pulse slice profiles. Therefore, in Flow-sensitive Alternating-Inversion Recovery (FAIR) imaging the slice selective (ss) inversion slab is usually three to four times thicker than the imaging slice. However, this reduces perfusion sensitivity due to the increased transit delay of the incoming blood with unperturbed spins. In the present article, the dependence of the magnetization on the RF pulse slice profiles is inspected both theoretically and experimentally. A perfusion quantification model is presented that allows the use of thinner ss inversion slabs by taking into account the offset of RF slice profiles between ss and nonselective inversion slabs. This model was tested in both phantom and human studies. Magn Reson Med 46:193-197, 2001.

Separation and quantification of perfusion and BOLD effects by simultaneous acquisition of functional I(0)- and T2(*)-parameter maps.

The nature of the coupling between neuronal activity and the hemodynamic response is the subject of intensive research. As a means to simultaneously measure parametric changes of T2(*), initial intensity (I(0)) and perfusion with high temporal resolution, a multi-image EPI technique with slice-selective inversion recovery (ssIR) for arterial spin labeling was developed and implemented. Comparative measurements with and without the preceding slice-selective inversion pulse were performed. I(0) and R2(*) changes induced by primary visual stimulation were separated. For ssIR-multi-image EPI the average change of I(0) over all 12 subjects was 3.4%, corresponding to a perfusion change of 40 ml/min/100 g, whereas only minor I(0) changes were observed without inversion. On average, the R2(*) of the activated pixels changed by -0.62 sec(-1) without inversion, while a significantly reduced average R2(*) change of -0.46 sec(-1) was calculated for ssIR-multi-image EPI due to a decreased BOLD effect contribution of the intravascular compartment.

Simultaneous detection of changes in perfusion and BOLD contrast

A new functional magnetic resonance imaging (fMRI) technique for simultaneous detection (SIDE) of changes in perfusion and blood oxygenation level dependent (BOLD) contrast is described. Perfusion contrast is generated by using magnetically labeled endogenous water proton spins as a freely diffusible tracer. A single slice-selective inversion pulse is combined with dual echo echo-planar imaging to generate a spin-echo (SE) image sensitive to changes in perfusion and a gradient-echo (GE) image sensitive to changes in both perfusion and BOLD contrast. The SIDE technique was applied to detect functional changes induced by a visual search task. A theoretical analysis is provided to calculate quantitative maps of changes in cerebral blood flow (DeltaCBF) and effective transverse relaxation time (DeltaT(2)*) from the corresponding signal changes in the SE and GE images. Since SE an GE images are generated from the same longitudinal magnetization, no errors due to spatial or temporal mismatch can arise in the quantification of DeltaCBF and DeltaT(2)*.